Predicting the severity of a disease outbreak is an important task for health personnel and college administrators. Influenza is a disease that is commonly transmitted amongst college students. While traditional methods of mathematical prediction utilize systems of differential equations to predict results at the macro level that can be compared readily to historical disease data, agent-based models attempt to detail individual interactions on a daily basis. Agent-based models contain independent agents that follow a few given rules, so there is room for change and experimentation that would be difficult with the traditional mathematical models. Through these rules, one hopes to discover the underlying behavior of the system at hand. Specifically, how disease spreads throughout a population of these agents, given some number of initially infected agents. Once parameters are established so that normal runs produce results consistent with the mathematical models, the agent-based model can be modified, so different results can be seen from different behaviors. Here, we create an agent-based model of influenza with a population of roughly 2000 agents, and measure the effectiveness of two simple methods of lessening the severity of an outbreak.


Pasteur, Drew

Second Advisor

Visa, Sophie


Computer Science; Mathematics


Infectious Disease | Numerical Analysis and Scientific Computing

Publication Date


Degree Granted

Bachelor of Arts

Document Type

Senior Independent Study Thesis



© Copyright 2013 Matthew A. Lambert